基于金字塔LK光流法及超声传感的微型无人机障碍识别优化研究

Research on obstacle identification optimization of micro UAV based on pyramid LK optical flow method and ultrasonic sensor

针对在有雾或能见度低的情况下,传统的光流算法在微型无人机障碍物辨识过程中准确性低、适应性差等问题,采用了一种基于金字塔LK光流法结合超声传感的避障方法。主要进行金字塔分布式检测,经超声波传感器探测目标,用测得的光流数据与IMU值融合后进行Kalman滤波处理,与设定的阈值进行比较,判断是否为障碍物,以提高微小型无人机障碍物辨识准确性。通过搭建无人机试验平台,对实验结果进行比较、分析,验证了方法的可行性。

In the case of fog or low visibility,an improved method based on pyramid LK optical flow method and ultrasonic sensor is adopted to solve the problems of low accuracy and poor adaptability of traditional optical flow algorithm in the process of obstacle identification of micro UAV.It mainly carries out pyramid distributed detection,detects the target by ultrasonic sensor,carries on Kalman filtering after fusion of measured optical flow data with IMU value,compares it with the set threshold,and judges whether it is an obstacle or not.This method can improve the accuracy of small UAV obstacle identification.By setting up the UAV test platform,the experimental results are compared and analyzed,and the feasibility of the algorithm is verified.